A comparative study of blood chemistry profile was conducted on two mullet (genus Mugil) populations from the Adriatic and Tyrrhenian Sea, which lived under different abiotic and biotic conditions. Specimens for the analysis have been caught on locations without any aquaculture activities. In both of analyzed groups, from the Adriatic (AM) and Tyrrhenian Sea (TM), the following parameters were monitored: blood plasma enzymes - aspartate and alanin aminotransferase (AST, ALT) and metabolites - triglycerides (TRIG); cholesterol (CHOL); glucose (GLU) and total proteins (TP). Significant difference was determined between biochemical parameters of two analyzed groups. Measured blood chemistry parameters were proved as good indicators of living conditions in different habitats. Classical statistical approaches were used for determination of dissimilarity in blood chemistry in relation to living conditions in different habitats. Machine learning technique was applied to generate classification model, and to find the importance, strength, mutual interactions or dependencies in analyzed blood chemistry parameters in the model, as well as to investigate reliability of particular parameters within the groups.
Blood biochemical approach in differentiation of Adriatic and Tyrrhenian mullet populations (Genus Mugil Linnaeus, 1758)
FAZIO, Francesco;PICCIONE, Giuseppe;FAGGIO, Caterina
2013-01-01
Abstract
A comparative study of blood chemistry profile was conducted on two mullet (genus Mugil) populations from the Adriatic and Tyrrhenian Sea, which lived under different abiotic and biotic conditions. Specimens for the analysis have been caught on locations without any aquaculture activities. In both of analyzed groups, from the Adriatic (AM) and Tyrrhenian Sea (TM), the following parameters were monitored: blood plasma enzymes - aspartate and alanin aminotransferase (AST, ALT) and metabolites - triglycerides (TRIG); cholesterol (CHOL); glucose (GLU) and total proteins (TP). Significant difference was determined between biochemical parameters of two analyzed groups. Measured blood chemistry parameters were proved as good indicators of living conditions in different habitats. Classical statistical approaches were used for determination of dissimilarity in blood chemistry in relation to living conditions in different habitats. Machine learning technique was applied to generate classification model, and to find the importance, strength, mutual interactions or dependencies in analyzed blood chemistry parameters in the model, as well as to investigate reliability of particular parameters within the groups.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.